> It could eventually make possible the kind of insight you describe (e.g. discovering a selection pressure that would flip X-risk reduction from undefined to positive).
Absolutely, although it could also lead people to mistakenly think they found such insight and do something that turns out bad for the long term. The crux then becomes whether we have any good reason to believe the former is determinately more likely than the latter, despite severe unawareness. To potentially make a convincing case that it is, one really needs to seriously engage with arguments for unawareness-driven cluelessness. Otherwise, one is always gonna give arguments that have already been debunked or that are way too vague for the crux to be identified, and the discussion never advances.
I take your point, and you’ve convinced me that determinative unawareness lies at a deeper level than the cluelessness my essay addresses, blocking any strong global justification for discovery’s net goodness.
Still, the local value of certain activities—surviving, cooperating, preserving options, and discovering—seems hard to ignore. If one continues to think the long term warrants attention and effort, even without clear awareness of how to help, discovery remains a prudent direction.
In that sense, I see your essay as defining the upper bound of what we can justifiably claim to know, and mine as exploring what orientation remains coherent within that bound. Any such orientation can only ever be conditionally coherent, but refining our map is still a principled way to reduce arbitrariness and local cluelessness, even if we can never know absolutely whether mapping is a net good.
Thanks for the reply :)
> It could eventually make possible the kind of insight you describe (e.g. discovering a selection pressure that would flip X-risk reduction from undefined to positive).
Absolutely, although it could also lead people to mistakenly think they found such insight and do something that turns out bad for the long term. The crux then becomes whether we have any good reason to believe the former is determinately more likely than the latter, despite severe unawareness. To potentially make a convincing case that it is, one really needs to seriously engage with arguments for unawareness-driven cluelessness. Otherwise, one is always gonna give arguments that have already been debunked or that are way too vague for the crux to be identified, and the discussion never advances.
I take your point, and you’ve convinced me that determinative unawareness lies at a deeper level than the cluelessness my essay addresses, blocking any strong global justification for discovery’s net goodness.
Still, the local value of certain activities—surviving, cooperating, preserving options, and discovering—seems hard to ignore. If one continues to think the long term warrants attention and effort, even without clear awareness of how to help, discovery remains a prudent direction.
In that sense, I see your essay as defining the upper bound of what we can justifiably claim to know, and mine as exploring what orientation remains coherent within that bound. Any such orientation can only ever be conditionally coherent, but refining our map is still a principled way to reduce arbitrariness and local cluelessness, even if we can never know absolutely whether mapping is a net good.